Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 204
D ATA - O R I E N T E D T R AV E L B E H AV I O R A N A LY S I S 193
recording omissions when people forget to do that. Such TABLE 1 Data to Be Acquired
recording omissions inevitably become noticeable when Data (Numerals Indicate Geographic
Points of Observation) Bytes
a survey is implemented for a longer period and in more
X-axis acceleration (32 Hz) 2
detail. However, recording detailed travel behavior for a Y-axis acceleration (32 Hz) 2
long period is indispensable for better understanding of Z-axis acceleration (32 Hz) 2
travel behavior and analyzing the dynamics of that Atmospheric pressure sensor (32 Hz) 2
Angular velocity (32 Hz) 2
behavior. To solve such problems, Hato (2006) proposed Ultraviolet ray (32 Hz) 2
a method for identifying travel activities by using infor- Direction (32 Hz) 2
mation from multiple sensors, with the aim of enabling Sound (10 Hz) 2
PS location (latitude, longitude, altitude, velocity,
the achievement of long-term observations by completely direction) (1 Hz) 23
eliminating the act of recording by subjects. Elliptical error of GPS location measurement 15
Hato has already developed a small, portable travel- NOTE: 88-day continuous recording (battery duration: about 3 days).
activity measuring instrument that requires no entry by OS is TRON (activity identification programs can be embedded or
subjects. Conventional surveys have collected identifica- rewritten in C).
tion information such as facility type, transport mode,
and activity content through the operation of instru- a record of acceleration variability of subjects in coffee
ments, questionnaires, and the like. However, these com- shops and CD shops. It shows that, in coffee shops, sub-
plicated surveys burden the subjects and rely on their jects move only when the menu is given by a waiter or
memory, problems often leading to recording omissions waitress or when they try to drink water in a cup, and
or incorrect records. Hato proposed a method for esti- the accelerations at these occasions have been recorded.
mating behavioral contexts by using behavioral-context In contrast, in CD shops, subjects often move around
addressable loggers in the shell (BCALS), a wearable, looking for CDs, and the variability in acceleration has
behavioral-context information-measuring instrument, been observed. Most of the cases that show no variabil-
for reestimating label information, such as facility type ity in acceleration probably indicate that such actions
and transport mode, from ecological and environmental include listening to a CD at a set location or paying at
sensors that are based on learning models. Figure 4 the cash register. Thus, it is possible to record detailed
shows the BCALS used in the present study, and Table 1 behaviors of subjects.
lists the data to be acquired. Acceleration information is Furthermore, Figure 7 shows changes in atmospheric
used for identifying the transport mode. Atmospheric pressure. Every time a subject changes floors or visits a
pressure is used in combination with ultraviolet rays for different facility, the atmospheric pressure changes con-
judging the floor level and whether the person is indoors siderably. The floor of an activity can be identified from
or outdoors. Sound and temperature are used for identi- data on atmospheric pressure.
fying the behavior content. A small logger equipped with multiple sensors has
Here are some examples of measurement results from been introduced. Acceleration and sound are effective
the sensors. Figure 5 shows the changes in acceleration for identifying activity content and location. (They also
of each transport mode. Walking has the largest variabil- enable capturing the number of steps and are effective
ity in acceleration and is followed by bicycling, motor- for evaluating walking environments.) Moreover, atmos-
bike, and automobile. It shows the possibility of
identifying transport modes on the basis of the magni-
tude of acceleration without asking subjects. Figure 6 is
FIGURE 5 Measurement results of changes in acceleration
FIGURE 4 Exterior view of BCALS. and noise.
